S. Haeghebaert ¹, M. Servat ², C. Duchen ³, J.C. Minet ⁴, A.E Agrech ⁴, I. Thièse ⁴, C. Leclerc ⁷, V. Vaillant ¹, C. Hemery ³ , E. Maillot ¹, C. Soulé ⁵, E. Pozio ⁶, P. Massip ⁷, J.F. Magnaval ⁷, J.C. Desenclos ¹
1.Réseau National de Santé Publique, Saint-Maurice, France
2. Direction Départementale des Affaires Sanitaires et Sociales du Tarn et Garonne, France
3. Cellule Interrégionale d’Epidémiologie d’Intervention du Sud-Ouest, France
4. Services Vétérinaires du Tarn et Garonne, France
5. Centre National d’Etudes Vétérinaires et Alimentaires, Maisons Alfort, France
6. Trichinella International Reference Centre, Istituto Superiore di Sanità, Rome, Italy
7. Centre Hospitalier Régional Universitaire, Purpan, Toulouse, France

Two outbreaks of trichinellosis in the Tarn et Garonne département, France, were reported by the departmental health autorities on 2 March 1998, to the Réseau National de Santé Publique (RNSP).

An epidemiological investigation began on 3 March in order to determine the extent of the outbreak, identify its source and food vector, and to propose control measures.

A case was defined as a resident of the Tarn, Tarn et Garonne or Haute-Garonne départements of the Midi-Pyrénées region who had presented with the following features since 1 January 1998:

confirmed case: fever (>38°C) with myalgia or facial oedema associated with a trichinella positive serology or muscle biopsy

probable case: at least three out of the following four criteria: fever, myalgia, facial oedema, hypereosinophilia >1000/mm³

suspected case: - hypereosinophilia >1000/mm³ alone or associated with fever or myalgia

- hypereosinophilia > 300/mm³ and < 1000/mm³ or an increase in creatinine phosphokinase (CPK), occurring in a family context of confirmed or probable trichinosis.

An active search for cases was conducted by the Directions Départementales des Affaires Sanitaires et Sociales (DDASS) and Cellule Interrégionale d'Epidémiologie d'Intervention du Sud-Ouest (CIREI) among medical laboratories, general practitioners, and hospital physicians in the Tarn, Tarn et Garonne, and Haute-Garonne départements. They were asked to report hypereosinophilia screening tests >1000/mm³ and patients who had consulted for symptoms suggestive of trichinellosis since 1 January.

Serodiagnostic tests for trichinellosis were performed by several specialised laboratories using enzyme-linked immunosorbant assay (ELISA) (Trichinella Serology Microtiter Elisa Kit, LMD Laboratories, Inc.). No muscle biopsy was performed.

A descriptive study was conducted on 5 March. A standardised questionnaire on the clinical features, date of onset of symptoms, laboratory tests performed, consumption of meat products, and dates and places of meat purchase during January and February was administered by telephone to the identified cases who could be contacted at the time of the study.

A case control study was conducted on 6 March to test the hypotheses generated by the descriptive study. Only confirmed or probable cases were included. When several members of the same family were sick, only one case was selected at random. Two controls per case were selected at random from telephone directories for the districts of residence of cases.

The data were analysed with Epi Info software, version 6.02 (CDC, Atlanta). The strength of the association between the factor studied and the disease was determined by the odds ratio (OR; maximum likelihood method). The stability of the association was tested using the Mantel-Haenszel test. The precision of the odds ratio was expressed by the 95% confidence interval (CI; maximum likelihood method).

A veterinary investigation was conducted by the departmental veterinary services, in parallel with the case finding and descriptive studies. Food sampling was performed according to findings of the epidemiological study in which cases were asked whether they had kept meat bought in January or early February in the freezer. The distribution channels were identified from the purchasing sites reported by the cases.

The food samples were tested using the enzymatic digestion method at the Laboratoire Vétérinaire Départemental (LVD, departmental veterinary laboratory), Montauban, and at the Centre National d’Etudes Vétérinaires et Alimentaires (CNEVA, national centre for veterinary and food studies), Maisons Alfort, France. Larvae isolated by trypsin digestion were typed by the Trichinella International Reference Centre in Rome, Italy (Dr Pozio), using genomic analysis methods (polymerase chain reaction and ramdom amplified polymorphism DNA).

A total of 128 cases were identified in three départements of the Midi-Pyrénées region, France, 79 of whom were confirmed cases (62%), 32 probable cases (25%), and 17 suspected cases (13%). No severe cases were reported. Six cases were admitted to hospital for antiparasitic treatment.

The epidemic curve (figure 1) suggests a common and point source of contamination in the 4th week of January. The case control study conducted on 17 cases and 34 controls showed that the risk of richinella infection was significantly higher in those who ate horse meat than in those who did not (OR = 14 ; 95% CI 3.1-82.5 ; p <0.01) (table 1).

Table1 : Frequency of exposure and strength of association. Trichinellosis outbreak, Midi-Pyrénées Region, January-March 1998

Food

Frequency of consumption

OR

95% CI

p

Cases

Controls

N = 17

N = 34

N

%

N

%

Pork

13

81%

28

85%

0.8

0.1 – 5.8

NS

Wild boar

2

12.5%

4

12%

1.1

0.1 – 8.5

NS

Beef

14

93%

32

94%

0.9

0.04 – 55.3

NS

Mutton or lamb

9

56%

13

39%

1.9

0.5 – 7.9

NS

Horse

12

76.5%

5

17.6%

14.1

3.1 – 82.5

P < 0.01

Delicatessen meats

12

86%

28

85%

1.1

0.1 – 12.7

NS

Larvae of Trichinella spiralis were found in a joint of horse meat bought by a case and kept in the freezer. Analyses by the LVD and CNEVA showed a low level of contamination (< 1 larva per 5 g of horse meat).

The inquiry into the supply and distribution channels identified a horse carcass from a batch of horses imported from the Federal Republic of Yugoslavia on 19 January and slaughtered in France. The meat was marketed in the week 19 to 25 January.

Cases were distributed geographically over the same area as the sites in the three départements of Tarn et Garonne, Haute-Garonne, and Tarn where meat from the carcass was sold.

The results of the epidemiological, parasitological, and veterinary investigations suggested that this outbreak was linked to the consumption of horse meat imported from the Federal Republic of Yugoslavia. The small number of cases associated with a whole carcass (264 kg of meat) and the absence of severe clinical illness are probably related to low levels of parasite contamination in the meat eaten by the cases.

This outbreak is the seventh community epidemic in France reported since 1976 and linked to the consumption of imported horse meat. The identification of trichinella larvae in the meat of the horse implicated by the epidemiological study enabled the epidemiological hypothesis suggested in the previous outbreaks to be confirmed, for the first time (1,2).

This epidemic shows the limits of the protocols currently used in the systematic control of horse carcasses after slaughtering (3), and suggests that increased vigilance is required with respect to horse meat, particularly when imported from eastern Europe.

Recent epidemiological bulletins from the Croatian National Institute of Public Health (4) showed a much higher incidence of human trichinellosis in January and February 1998 than in previous years. Moreover, an outbreak of trichinellosis occurred in Piacenza, in Northern Italy at the same time as the French outbreak. This outbreak was shown to be associated with the consumption of the head of a horse imported from eastern Europe, which was mistakenly released on the market despite a trichinella-positive control (5).

1. Ancelle T, Dupouy-Camet J, Desenclos JC, Maillot E, Charlet F, Gravelat-Desclaux C et al. Epidémie de trichinellose (France, 1993). Bilan des investigations. Bulletin Epidemiologique Hebdomadaire 1994; 29: 127-129.

2. Ancelle T. History of trichinellosis outbreaks linked to horse meat consumption, 1975-1998. EuroSurveillance 1998; 3: 86-9

3. Maillot E. Trichinellosis associated with horse meat consumption: European regulations and risk management. EuroSurveillance 1998; 3: 90

4. Epidemiological News. Croatian National Institute of Public Health, 1998 (1,2)

5. Pozio E. Human outbreak of trichinellosis associated with the consumption of horsemeat in Italy. EuroSurveillance 1998; 3: 85-6